UBC Faculty Research and Publications

A systematic experimental study on gold recovery from electronic waste using selective ammonium persulfate… Alzate, Andrea; López, Esperanza; Serna, Claudia; Holuszko, M. E.; Gonzalez, Oberlando Apr 30, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-Alzate_A_et_al_Systematic_experiemental_CoM_2017.pdf [ 2.22MB ]
Metadata
JSON: 52383-1.0347214.json
JSON-LD: 52383-1.0347214-ld.json
RDF/XML (Pretty): 52383-1.0347214-rdf.xml
RDF/JSON: 52383-1.0347214-rdf.json
Turtle: 52383-1.0347214-turtle.txt
N-Triples: 52383-1.0347214-rdf-ntriples.txt
Original Record: 52383-1.0347214-source.json
Full Text
52383-1.0347214-fulltext.txt
Citation
52383-1.0347214.ris

Full Text

International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017   A SYSTEMATIC EXPERIMENTAL STUDY ON GOLD RECOVERY FROM ELECTRONIC WASTE USING SELECTIVE AMMONIUM PERSULFATE OXIDATION  Andrea Alzate*1, Esperanza López1, Claudia Serna1, Maria Holuszko2, Oberlando Gonzalez3  1GIPIMME Research Group, Department of Materials Engineering, University of Antioquia, CL 67 53-108, Medellín-Colombia (*Corresponding author: andreaalzatenaranjo@gmail.com)  2NBK Institute of Mining Engineering, University of British Columbia, 517-6350 Stores Road, Vancouver, BC, V6T 1Z4, Canada  3Ingeniería, Suministros y Montajes S.A.S, INSUMON S.A.S, CL 36 36-9, Medellín-Colombia  ABSTRACT  This paper presents a systematic approach on gold recovery from electronic waste (e-waste) using ammonium persulfate. This process was developed as a response to the lack of hydrometallurgical systems capable of separating gold from its metallic substrate without material grinding, with minimum formation of pollutants, and achieving a lower reaction time. Computer memory boards, electronic processors and electronic pins and contacts were analyzed to determinate base metals (Ni, Fe, Cu) and gold (Au) using microwave plasma atomic emission spectroscopy (MP-AES). An aqueous commercial grade ammonium persulfate with oxygen and pressure was used to produce the persulfate anion (S2O82-) and the oxidative sulfate ion (SO42-), which partially oxidized and leached the base metals breaking the Au-Ni-Fe-Cu bonds and allowing gold to be recovered in its original non-leaching state. The influence of the oxidative parameters was evaluated using full factorial (FF) and central composite designs (CCD) with response surface methodology (RSM) and first and second order models were developed. Using RSM allowed to obtain a faster recovery of gold, minimizing the agent consumption. The findings presented suggest that optimized quantities of ammonium persulfate, oxygen, pressure, temperature and liquid solid ratio could be used for selective oxidization of the base metals and to extract more than 96% of gold from e-waste.     KEYWORDS  Electronic waste, Experimental design, Gold recovery, Persulfate, Response surface methodology (RSM), Selective oxidation   International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  INTRODUCTION  The rapid growth in the manufacturing of technological devices has generated massive quantities of electronic waste (e-waste). The e-waste valuable metals and hazardous elements and the environmental issue of inadequate disposal have increased the interest for developing sustainable extraction techniques to permit the recovery of metals and the proper treatment of the hazardous materials. Large quantities of precious metals (Au, Pd, Ag) and base metals (Cu, Ni, Fe) are found in e-waste (Sanyal et al., 2013; Yazici and Deveci, 2014). They are usually distributed in the epoxy resin copper-clad laminate structure and in the other type of metallic materials used in the manufacturing electronic parts (Ghosh et al., 2015). Studies of the morphological features of the printed circuit boards (PCB) of discarded mobile phones and electronic pins found in central processing units (CPU) show structures divided in layers, including a superficial metallic fraction made of a gold-nickel alloy and internal base metals deposited in plastic or metallic substrates (Barbieri et al., 2010; Ha et al., 2014). The metallic fraction content in PCB is about 28-30%, specifically 10-20% cooper, 1-5% lead, 1-3% nickel, and 0.3-0.4% precious metals (Sarvar et al., 2015).  The recovery of metallic fraction from e-waste has been performed using conventional hydrometallurgical methods. The common focus of these methods is the primary shredding, crushing or grinding of the material, followed by the total leaching of the metals of interest (Tuncuk et al., 2012). Numerous leaching studies have used strong acids (HCl, HNO3, H2SO4) and oxidative reagents (cyanide, thiourea, halide, nitrate, iodide, and thiosulfates) to recover silver, gold and palladium from e-waste (Naseri Joda and Rashchi, 2012; Xiu et al., 2015; Zhang et al., 2012). These types of procedures are time consuming and have to include many stages to reach the total metal dissolution and the transformation from the soluble to the elemental recoverable form. For instance, leaching of gold and silver from e-waste with cyanide, nitric acid and thiosulfate has been performed with reaction times greater than 2 hours (Petter et al., 2014). Furthermore, extensive stages of purification such as cementation, solvent extraction, precipitation and coagulation, have been reported to reach be efficient in the recovery of gold from secondary sources (Syed, 2012). Another concern of the conventional leaching processes is the extensive use of cyanide, thiourea, halides and some strong acids that are recognized by their toxic potential, low chemical stability, and environmental problems due to inadequate handling (Tuncuk et al., 2012). Consequently, developing new hydrometallurgical processes for e-waste management should be a priority in research to meet the current challenges in the application of new non-pollutant, non-toxic agents as well as the reduction of reaction time to be more economical  (Alzate et al., 2016a).  In recent years, some processes that attempted to use environmentally friendly agents and the optimization of gold and base metals recovery from e-waste have been suggested  (Alzate et al., 2016a; Barbieri et al., 2010; Ha et al., 2014; Syed, 2006). For instance, alternative agents to extract non-leaching gold that include ammonium persulfate (NH4)2S2O8), potassium persulfate (K2S2O8), cupric chloride (CuCl2) and persulfate salt-ammonia were studied to recover gold and silver from e-waste (Alzate et al., 2016a; Barbieri et al., 2010; Hyk & Kitka, 2017; Syed, 2012). These agents were used to oxidize and leach the metal substrate (Ni, Fe, and Cu) where gold and silver were superficially associated as coating (Barbieri et al., 2010). The partial leaching of the substrate permitted gold recovery in a solid and particulate state. Due to the elimination of gold leaching, it was possible to reduce reaction time, avoiding purification stages and achieving the 98% in Au recovery with a minimum formation of contaminant by-products or total agent regeneration (Alzate et al., 2016a; Barbieri et al., 2010; Syed, 2006).  Selective oxidation by persulfate can be optimized in order to maximize the recovery of gold from e-waste reducing parameters like agent consumption and reaction time (Birloaga et al., 2013; Hadi et al., 2015; Jordão et al., 2016). A way to optimize this process is to use response surface methodology (RSM). This methodology is currently used for the optimization of processes in which the response of interest is influenced by different parameters (Montgomery, 2013; Bas and Boyaci 2007). RSM was reported in e-waste management to determine the leaching of gold, evaluating the incidence of thiosulfate, cooper and ammonia concentration (Ha et al., 2014). In addition, the extraction of Cu, Fe, Ni, Ag and Pd from PCB was studied, adopting RSM to establish the conditions maximizing metals extraction using H2SO4-CuSO4-NaCl solutions (Yazici & Deveci, 2013). International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  In previous work we developed environmentally friendly systems to extract non-leaching gold by partial base metals oxidation from e-waste using ammonium persulfate (Alzate et al., 2016a, 2016b, 2015). The concentration, oxygen flow, pressure, temperature and L/S ratio were analyzed to optimize the base metals oxidation and gold recovery and the conditions for each system were established. In the present paper, a comparative experimental study for the previously developed systems is presented. The effects of (NH4)2S2O8 concentration (0.22-1.31 M), oxygen flow (0.0-1.4 L/min), pressure (101-203 kPa), temperature (100-137°C) and liquid/solid ratio (10-30 mL/g) over the recovery of gold were analyzed using full factorial 23 and central composite designs (CCD) with response surface methodology (RSM). First and second order models were developed for the prediction of Au recovery from e-waste and quantities greater than 96% of Au were recovered. This study aimed at maximizing Au recovery with a novel methodology that includes the selective oxidation of base metals with ammonium persulfate and the optimization of the most significant parameters.   EXPERIMENTAL  Materials and reagents  The computer memory boards (sample 1), electronic processors (sample 2) and electronic pins and contacts (sample 3) (Figure 1) used in this study were obtained from the Colombian companies INSUMON S.A.S and LITO S.A.S. A total amount of 50 end-of-life computer memory boards, 50 end-of-life electronic processors and 500g of electronic pins and contacts were separately classified and used for chemical characterization, selective base metals oxidation, and gold recovery tests without grinding stages.              Figure 1. (a) Sample 1. Computer memory boards (b) Sample 2. Electronic processors (c) Sample 3. Electronic pins and contacts  Quantities of 35.00 g of processors, 48.00 g of memory boards and 20.06 g of electronic pins and contacts were used to determine the amount of gold (Au) and base metals (Fe, Ni, Cu) by chemical digestion using aqua-regia (Lee et al., 2011; Petter et al., 2014) followed by microwave plasma atomic emission spectroscopy (MP-AES, AGILENT MP 4100) (Table 1). After the chemical characterization, aqueous commercial grade ammonium persulfate (≥98% (NH4)2S2O8) with a water solubility of 850 g/L at 25 °C (Hernandez, 2005) and 6.98% of active oxygen (Turan et al., 2015) was the selected eco-friendly agent used to produce the persulfate anion (S2O82-). This anion is not absorbed or bio accumulated in soil due to its high water solubility and fast dissociation (Hernandez, 2005). Additionally, the S2O82- can be easily activated in aqueous solutions by increasing temperature to produce the oxidative sulfate ion (SO42-) that is considered as an inert and non-pollutant product (Sharma et al., 2015). The produced SO42- generated a system of oxidative reactions that partially leached the base metals breaking the Au-Ni-Fe-Cu bond and allowing gold to be extracted in its original non-leaching state. The concentration of (NH4)2S2O8, oxygen flow, pressure, temperature and liquid/solid ratio over the partial base metals oxidation and gold (a) (b) (c) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  recovery was studied using a systematic approach to the experimental designs.  Table 1. Chemical composition for the electronic waste E-waste sample Metal content (g/kg) Cu Fe Ni Au Ag Sample 1. Computer memory boards  69.97 9.76 8.74 0.62 <0.001 Sample 2. Electronic Processors 26.65 86.00 73.64 1.05 <0.001 Sample 3. Electronic pins and contacts 972.79 0.09 4.64 5.90 <0.001  Selective ammonium persulfate oxidation and gold recovery  The selective oxidation of the base metals (Ni, Fe and Cu) and the recovery of gold were performed in the samples 1, 2 and 3 using the selected parameters: (NH4)2S2O8 concentration (0.22-1.31 M), oxygen flow (0.0-1.4 L/min), pressure (101-203 kPa), temperature (100-137°C) and liquid/solid ratio (10-30 mL/g). Solutions were prepared in glass reactors with deionized water by heating at boiling point. Mechanical stirring was used to reach complete dissolution and speciation of persulfate (S2O8)2- in the terms of availability of an oxidative anion (SO42-) (Alzate et al., 2016a). Samples 1, 2 and 3 were separately put into the solutions, while oxygen or pressure were introduced into the systems to maximize base metals oxidation and to reduce reaction time. During the reaction periods, iron, nickel and copper were partially oxidized and these elements were analyzed using atomic absorption spectroscopy (AAS, THERMO S4), microwave plasma atomic emission spectroscopy (MP-AES, AGILENT MP 4100) and X-Ray fluorescence spectroscopy (XRF). The selective base metals oxidation led to the breakage of the Au-Ni-Fe-Cu bonds and gold was released, hence recovered in its original solid state. Recovery of gold was calculated based on the initial gold content in the electronic waste and the released gold from each sample was determined at optimum conditions. The concentration of (NH4)2S2O8, oxygen pressure, temperature and liquid/solid ratio on the recovery of gold were analyzed using a systematic experimental design and developed mathematical models. The chemical composition and morphology of the recovered gold was analyzed by SEM with energy dispersive X-ray spectroscopy (SEM/EDX, JEOL JSM-6490LV).  Systematic experimental design  Studying the base metals oxidation and gold recovery from samples 1, 2 and 3 were performed by adopting sequential experimental designs (Montgomery, 2013) and analyzed using Design Expert software (2015). Linear effects were evaluated adopting full factorial 23 experimental designs in which first order models were established using the regression linear model of three factors with fixed effects (Eq.1) (Montgomery, 2013; Montgomery et al., 2011). Eq. (1) was used to predict the behavior of the response (y) related to the calculated linear (β1, β2, β3) and interaction (β12, β13, β23, β123) coefficients. First order models were tested for the samples 1, 2 and 3. However, the application of these models for samples 2 and 3 showed significant lack of fit and second order models were developed to adjust the first order model to the curvature.   y =  β0 + β1x1 + β2x2 +  β3x3 + β12x1x2 +  β13x1x3 + β23x2x3 +  β123x1x2x3                   (1)  After evaluating the adequacy of the first order models, second order models were proposed with the aim to fit the data and to optimize the test conditions for the recovery of gold from samples 2 and 3. Response surface methodology (RSM) is the statistical technique used to improve and optimize the test conditions for the recovery of gold. This methodology uses a second order polynomial model (Eq.2) that includes the quadratic coefficients (β11, β22 and β33) (Montgomery, 2013; Bas et al., 2007; Walpole et al., 2012).   y = β0 + β1x1 + β2x2 + β3x3 + β11x12 + β22x22 + β33x32 + β12x1x2 + β13x1x3 + β23x2x3          (2)  The adjustment to the second order model was performed using a central composite design (CCD) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  where the number of runs was given by (2k+2k+N0) with k is the number of factors, 2k is the starting factorial design, and N0 is the number of experiments at the center. The full factorial 23 experimental design for sample 1 is shown in table 2 while the central composite designs for the samples 2 and 3 are presented in tables 3 and 4.  Table 2. Full factorial 23 Experimental design for the recovery of gold from sample 1 Factors Coded and actual levels  Low (-1) High (+1) A: [(NH4)2S2O8] (M) 0.66 1.31 B: Pressure (kPa) 101 203 C: Liquid/solid Ratio (mL/g) 15 25  Table 3. Central composite design (CCD) for the recovery of gold from sample 2 Factors  Coded and actual Levels   -α (-1.682) (-1) (0) (+1) +α (+1.682) A: [(NH4)2S2O8] (M) 0.22 0.44 0.66 0.88 1.10 B: Oxygen (L/min) 0.0 0.2 0.6 1.0 1.4 C: Liquid/Solid Ratio (mL/g) 10 15 20 25 30  Table 4. Central composite design (CCD) for the recovery of gold from sample 3 Factors  Coded and actual Levels   -α (-1.682) (-1) (0) (+1) +α (+1.682) A: [(NH4)2S2O8] (M) 0.51 0.66 0.88 1.10 1.25 B: Temperature (°C) 100 110 120 130 137 C: Liquid/Solid Ratio (mL/g) 12 15 20 25 28  RESULTS AND DISCUSSION  Recovery of gold from computer memory boards  The effects of (NH4)2S2O8 concentration, oxygen pressure and liquid/solid ratio on the recovery of gold from computer memory boards were analyzed running the 16 experiments of the full factorial 23 design. The experimental design layout with actual levels of each factor is shown in table 5. The standard deviation of the response (Recovery of gold) was calculated to be ≤ 0.73%. The maximum recovery of gold (99.36% ± 0.45) was reached at point (-1, +1, +1) in a reaction time of 5 minutes. In table 6, the analysis of variance ANOVA for the model is shown. All linear effects and interactions were statistically significant with p-values ≤ 0.05.  Table 5. Experimental design layout for Au recovery from computer memory boards Exp Factors with actual levels Responses N0 A B C Au recovery (%) Base metals Oxidation (%) Reaction time (min) [(NH4)2S2O8] (M) Pressure (kPa) L/S Ratio (mL/g) Ni Fe Cu 1-2 0.66 101 15 0.00 ± 0.00 77.53 ± 1.87 44.18 ± 3.16 46.14 ± 1.39 40 3-4 1.31 101 15 10.67 ± 0.95 38.48 ± 2.09 41.68 ± 0.54 19.24 ± 0.74 30 5-6 0.66 203 15 13.83 ± 0.35 69.79 ± 1.26 68.80 ± 1.53 34.86 ± 0.18 5 7-8 1.31 203 15 41.32 ± 0.47 21.54 ± 0.52 15.60 ± 0.54 21.5 ± 1.50 5 9-10 0.66 101 25 0.00 ± 0.00 21.36 ± 0.27 19.69 ± 0.60 30.15 ± 2.03 40 11-12 1.31 101 25 8.82 ± 1.34 22.37 ± 0.25 26.75 ± 0.43 11.35 ± 1.91 20 13-14 0.66 203 25 99.36 ± 0.45 61.05 ± 0.10 76.63 ± 1.28 24.29 ± 1.22 5 15-16 1.31 203 25 98.79 ± 1.03 63.50 ± 0.95 71.46 ± 0.55 14.93 ± 0.10 5 International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017   Table 6. Analysis of variance (ANOVA) for Au recovery from computer memory boards Source Sum of Squares Degree of freedom Mean Squares p-value Model 24838.82 7 3548.40 < 0.0001 A- [(NH4)2S2O8] 538.36 1 538.36 < 0.0001 B- Pressure 13665.03 1 13665.03 < 0.0001 C- L/S Ratio 4980.48 1 4980.48 < 0.0001 AB 13.82 1 13.82    0.0009 AC 223.73 1 223.73 < 0.0001 BC 5245.74 1 5245.74 < 0.0001 ABC 171.68 1 171.68 < 0.0001 Pure Error 4.28 8 0.53  Corrected Total 24843.10 15    The regression model that describes the recovery of gold from sample 1 was established calculating the coefficients for each factor and developing a first order equation (Eq.3). The Predicted R-Squared of 0.9993 was in reasonable agreement with the adjusted R-Squared of 0.9997 with a difference ≤0.20, and the coefficient of determination (R2=0.9998) proved that the 99.98% of the variability in the response has been explained by the model.   Au recoverysample 1 = 34.10 + 5.80A + 29.22B + 17.64C + 0.93AB − 3.74AC + 18.11BC − 3.28ABC (3)  Equation 3 shows that the linear factors (A, B and C) and the interactions (AB) and (BC) had a positive effect on the recovery of gold while the (AC) interaction had a negative influence. Positive sign of A, B and C indicates that the increment of concentration, pressure and L/S ratio positively influenced the oxidation of the substrate allowing the release of gold in its solid state. This is explained by the S2O82-/activation with SO42- generation, the delivery of active oxygen from the agent, the formation of O2 by thermal decomposition, and the fast transfer and dissolution of oxygen in the aqueous system by the partial pressure increase. The fast oxidation of the metal substrate at 203kPa permitted the breakage of the Au-Ni-Fe-Cu bond without leaching of Au from the sample. Release of gold as solid particles was achieved after removing the sample from the reactor and applying a micro scale pressure washing. Figures 2a, 2b and 2c show that increasing pressure to the highest level permitted gold recovery greater than 99.36%. Maximum gold recovery was achieved under the (NH4)2S2O8 concentration ranging from 0.66 to 1.31 M, pressure of 203 kPa and L/S ratio of 25 mL/g.  Recovery of gold from electronic processors  The optimal release of solid gold from sample 2 using selective base metals oxidation was achieved when running seventeen experiments of the central composite design. The experiment design layout with the actual levels of each factor is shown in table 7. The standard deviation of the response (%Au recovery) was calculated to be ≤ 2.20%. The maximum recovery of gold (99.80%) was reached in the axial point (+α, 0, 0) where reagent concentration was at 1.10 M and oxygen was added in the central level of 0.6 L/min with an L/S ratio of 20 mL/g. In table 8, the ANOVA for the model is shown. All quadratic effects were statistically significant with p-values of <0.0001 at a confidence level of 95%. These results confirm the curvature associated to the model. Experiments at central levels showed high Au recoveries (≥98.86%) minimizing (NH4)2S2O8 concentration at 0.66 M and reducing the reaction time to 10 minutes. International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)99.670X1 = A: ConcentrationX2 = B: PressureActual FactorC: L/S Ratio = 250.66 0.79 0.92 1.05 1.18 1.31100120.6141.2161.8182.4203Au Recovery (%)A:Concentration (M)B:Pressure (kPa)2040608095Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)99.670X1 = A: ConcentrationX2 = C: L/S RatioActual FactorB: Pressure = 2030.66 0.79 0.92 1.05 1.18 1.31151719212325Au Recovery (%)A:Concentration (M)C:L/S Ratio (mL/g)2040608095(a) (b) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017   Figure 2. Contour plots for Au recovery from sample 1 (a) A: [(NH4)2S2O8] and B: Pressure with C=25 mL/g, (b) A: [(NH4)2S2O8] and C: L/S ratio whit B=203 kPa, (c) B: Pressure and C: L/S ratio with A=1.31M. Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)99.670X1 = B: PressureX2 = C: L/S RatioActual FactorA: Concentration = 1.31100 120.6 141.2 161.8 182.4 203151719212325Au Recovery (%)B:Pressure (kPa)C:L/S Ratio (mL/g)204060803015(c) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  Table 7. Experimental design layout for Au recovery from electronic processors. Exp Aleatory Factors with actual levels Base metals oxidation and Au recovery over 10 min N0 N0 A B C Cu (%) Fe (%) Ni (%) Au recovery (%) [(NH4)2S2O8] (M) Oxygen (L/min) L/S Ratio (mL/g) 1 3 0.44 0.2 15 1.74 31.70 6.42 73.61 2 15 0.88 0.2 15 1.94 10.80 17.94 81.69 3 7 0.44 1.0 15 4.21 24.50 74.41 88.81 4 1 0.88 1.0 15 1.72 4.82 4.55 98.22 5 6 0.44 0.2 25 2.42 31.50 9.73 86.82 6 16 0.88 0.2 25 3.66 42.33 7.86 94.61 7 12 0.44 1.0 25 2.79 40.07 43.03 88.85 8 4 0.88 1.0 25 5.00 44.49 3.37 99.11 9 17 0.22 0.6 20 2.91 4.33 23.79 74.09 10 9 1.10 0.6 20 5.54 6.79 21.66 99.80 11 11 0.66 0.0 20 15.12 47.57 64.75 72.98 12 13 0.66 2.0 20 2.86 22.48 24.59 97.91 13 2 0.66 0.6 10 2.57 6.77 6.72 76.38 14 14 0.66 0.6 30 3.32 2.95 30.53 98.97 15 10 0.66 0.6 20 3.53 35.15 9.85 99.39 16 5 0.66 0.6 20 3.79 31.53 9.06 98.86 17 8 0.66 0.6 20 3.60 37.48 10.31 99.26  International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  Table 8. Analysis of variance (ANOVA) for Au recovery from electronic processors Source Sum of Squares Df Mean Square p-value Model 2197.30 9 244.14 < 0.0001 A-[(NH4)2S2O8] 625.36 1 625.36 < 0.0001 B-Oxygen 665.39 1 665.39 < 0.0001 C-L/S Ratio 410.94 1 410.94 < 0.0001 AB 3.62 1 3.62 0.4011 AC 0.077 1 0.077 0.9014 BC 158.70 1 158.70 < 0.0001 A2 187.44 1 187.44 < 0.0001 B2 235.43 1 235.43 < 0.0001 C2 166.06 1 166.06 < 0.0001 Residual 72.69 15 4.85  Pure Error 8.74 10 0.87  Cor Total 2269.99 24     Contour plots were generated for the combination of three factors with actual values where the third factor was fixed at the central level.  Figures 3a and 3b show a wide red region in the middle of the operation surface where recoveries of gold were greater than 98%. This can be achieved under combined (NH4)2S2O8 concentration ranging from 0.66 to 1.10 M with Oxygen flow from 0.6 to 1.4 L/min and L/S ratio from 20 and 30 mL/g. The linear interaction between AB and AC did not present a significant effect on the response. This could be explained by the preferential substrate oxidation, the interaction of the quadratic factors and the loss of linearity in the model. Figure 3c shows that increasing the oxygen and L/S ratio up to the central levels with a fixed reagent concentration value of 0.66 M permitted the recovery of quantities of gold greater than 98.86%. International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)99.872.76X1 = A: [((NH4)2S2O8)]X2 = B: OxygenActual FactorC: L/S Ratio = 200.22 0.44 0.66 0.88 1.100.350.71.051.4Au Recovery (%)A: [(NH4)2S2O8] (M)B: Oxygen (L/min)70808090959898100Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)99.872.76X1 = A: [((NH4)2S2O8)]X2 = C: L/S RatioActual FactorB: Oxygen = 0.60.22 0.44 0.66 0.88 1.1101418222630Au Recovery (%)A: [(NH4)2S2O8] (M)C: L/S Ratio (mL/g)70808090959598100(a) (b) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017   Figure 3. Surface plots for Au recovery from sample 2 (a) A: [(NH4)2S2O8] and B: oxygen with C=20 mL/g, (b) A: [(NH4)2S2O8] and C: L/S ratio whit B=0.6 L/min, (c) B: oxygen and C: L/S ratio with A=0.66M.  The quadratic polynomial equation for Au recovery from sample 2 (Eq. 4) was developed in terms of coded factors. All the linear factors and the interaction AB and AC had a positive influence on the response while increasing the linear levels for B and C had a negative influence on the response as the BC interaction showed. The Predicted R-Squared of 0.86 was in reasonable agreement with the adjusted R-Squared of 0.95 with a difference ≤0.20, and the coefficient of determination (R2=0.97) proved that the 97% of the variability in the response has been explained by the model.   Au Recovery Sample 2 = 98.77+5.10A+5.27B+4.14C+0.48AB+0.069AC –3.15BC–3.11A2–3.48B2–2.92C2 (4)  Recovery of gold from electronic scrap pins and contacts  Oxidation of the base metals from the electronic scrap pins and contacts was run adopting all the possible combination between factors and real levels for the central composite design. The experiment design layout with actual levels of each factor is shown in Table 9.  The standard deviation for the main response (Au recovery) was calculated to be ≤ 4.43%. The maximum recovery of gold (96.78%) was reached at the center point (0, 0, 0) where (NH4)2S2O8 concentration was optimized at 0.88 M and temperature was increased until 120ºC with a L/S ratio of 20 mL/g. The base metals selective oxidation broke the bond Au-Ni and gold was released and recovered in its solid state in a reaction time of 15 minutes. Desirable levels of base metals oxidation can be achieved increasing temperature above 120ºC and combined (NH4)2S2O8 concentration in a range of 0.66-1.10M and L/S ratio between 20-28mL/g. In table 10, the ANOVA for the model is shown. The factors A, B, C, A2 and B2 and the interaction AB were statistically significant with p-values of <0.0001 at a confidence level of 95%.    Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)99.872.76X1 = B: OxygenX2 = C: L/S RatioActual FactorA: [((NH4)2S2O8)] = 0.660 0.35 0.7 1.05 1.4101418222630Au Recovery (%)B: Oxygen (L/min)C: L/S Ratio (mL/g)70809090959598100(c) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  Table 9. Experimental design layout for Au recovery from electronic pins and contacts Exp Aleatory Factors with actual levels Base metals oxidation and Au recovery over 15 min N0 N0 A B C Cu (%) Fe (%) Ni (%) Au recovery (%) [(NH4)2S2O8] (M) Temperature (°C) L/S ratio (mL/g) 1 12 0.66 110 15 1.34 0.00 11.23 39.10 2 9 1.10 110 15 1.89 0.01 14.76 68.28 3 13 0.66 130 15 1.40 0.02 26.81 85.96 4 6 1.10 130 15 2.07 0.03 23.11 82.02 5 7 0.66 110 25 2.38 52.50 18.73 57.21 6 5 1.10 110 25 3.77 0.00 16.77 80.59 7 10 0.66 130 25 2.78 0.01 56.39 91.31 8 11 1.10 130 25 4.21 0.02 48.26 95.47 9 1 0.51 120 20 0.01 19.99 8.22 62.31 10 3 1.25 120 20 0.09 30.24 41.30 93.23 11 16 0.88 103 20 0.00 21.55 10.11 36.25 12 14 0.88 137 20 2.88 0.00 39.55 95.97 13 8 0.88 120 12 0.00 0.00 25.71 84.43 14 2 0.88 120 28 0.17 99.00 58.24 96.48 15 17 0.88 120 20 0.00 25.02 30.32 96.78 16 15 0.88 120 20 0.00 23.27 30.10 95.21 17 4 0.88 120 20 0.00 30.94 26.33 96.69  Contour plots for gold recovery were generated for the combination of three factors with actual values where the third factor was fixed at the central level (Figure 4a, 4b and 4c). The results showed a wide red region in the operation surface where recoveries of gold were greater than 95%. This could be achieved under a combined (NH4)2S2O8 concentration of 0.88 M with a temperature from 120 to 130 °C and L/S ratios from 20 to 28 mL/g for a reaction time of 15 minutes. The maximum recovery of gold (96.78%) was reached at the center point (0, 0, 0) where the parameters were optimized. The regression model that describes the recovery of gold from electronic scrap pins and contacts by selective oxidation in (NH4)2S2O8 solutions was established by calculating the coefficients for each factor and developing a quadratic polynomial equation (Eq.5). Au Recovery Sample 3 = 96.36+7.67A+15.38B+5.09C-6.54AB+0.29AC –1.45BC–6.97A2–11.10B2–2.49C2   (5)   International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  Table 10. Analysis of variance (ANOVA) for Au recovery from electronic pins and contacts Source Sum of Squares df Mean Square p-value Model 6338.59 9 704.29 < 0.0001 A-[(NH4)2S2O8] 803.92 1 803.92 0.0004 B-Temperature 3229.66 1 3229.66 < 0.0001 C-L/S Ratio 353.54 1 353.54 0.0038 AB 342.43 1 342.43 0.0042 AC 0.66 1 0.66 0.8596 BC 16.88 1 16.88 0.3847 A2 548.29 1 548.29 0.0011 B2 1388.09 1 1388.09 < 0.0001 C2 69.85 1 69.85 0.1013 Residual 137.47 7 19.64  Pure Error 1.55 2 0.78  Cor Total 6476.06 16     Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)96.7836.25X1 = A: ConcentrationX2 = B: TemperatureActual FactorC: L/S Ratio = 200.66 0.77 0.88 0.99 1.1110115120125130Au Recovery (%)A: Concentration (M)B: Temperature (°C)5060708090100(a) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017    Figure 4. Contour plots for Au recovery from sample 3 (a) A: [(NH4)2S2O8] and B: Temperature with C=20 mL/g, (b) A: [(NH4)2S2O8] and C: L/S ratio whit B= 120°C, (c) B: Temperature and C: L/S ratio with A=0.88M.  Optimum conditions for the recovery of gold from e-waste samples  The optimum conditions for gold recovery from e-waste were established as follows: Sample 1; (NH4)2S2O8 concentration (0.66 M), pressure (203 kPa) and L/S Ratio (25 mL/g) for a reaction time of 5 min, Sample 2; (NH4)2S2O8 concentration (0.66 M), Oxygen flow rate (0.6 L/min) and L/S ratio (20 mL/g) Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)96.7836.25X1 = A: ConcentrationX2 = C: L/S RatioActual FactorB: Temperature = 1200.66 0.77 0.88 0.99 1.1151719212325Au Recovery (%)A: Concentration (M)C: L/S Ratio (g/mL)8090901009685Design-Expert® SoftwareFactor Coding: ActualAu Recovery (%)96.7836.25X1 = B: TemperatureX2 = C: L/S RatioActual FactorA: Concentration = 0.88110 115 120 125 130151719212325Au Recovery (%)B: Temperature (°C)C: L/S Ratio (g/mL)8090100857096(b) (c) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  for a reaction time of 10 min and, Sample 3; (NH4)2S2O8 concentration (0.88 M), Temperature (120ºC) and L/S ratio (20 mL/g) for a reaction time of 15 min. The time dependent release of gold from the samples is shown in figure 5. The Au recovery kinetics was strongly influenced by the increase of pressure, temperature and the addition of oxygen. The proposed systems for each sample demonstrated to be optimum alternatives to recover gold from e-waste by selective ammonium persulfate oxidation.   Figure 5. Au recovery kinetics from e-waste samples at optimum conditions  Releasing of gold coatings at optimum conditions was reached by the selective oxidization of the base metals. This oxidation broke the Au-Ni-Fe-Cu bonds and gold coatings were removed from the substrate by micro washing and filtration. The recovered Au was analyzed by SEM with energy dispersive X-ray spectroscopy (SEM/EDX) to evaluate morphology and chemical composition. Figures 6, 7 and 8 show the SEM/EDX analyses for the recovered Au from sample 1, 2 and 3 respectively. As it is shown in figure 6, 7 and 8, the fine gold coating was entirely recovered without evidence of having been corroded by (NH4)2S2O8. These findings can be compared with those obtained by Barbieri et al (2010) using CuCl2 and Syed (2006) using K2S2O8. Figures 6b, 7b and 8b show the grade of purity for the removed Au at the optimized levels. Higher peaks in Figure 6b, 7b and 8b correspond to the main energetic levels of gold that are presented at 2.12 keV and 9.71 keV.         0204060801000 2 5 6 8 10 15 30 40Au recovery  (%) Time (Minutes)  Sample 1 (0.66 M, 203 kPa, 25 mL/g)Sample 2 (0.66 M, 0.6 L/min, 20 mL/g) Sample 3 (0.88 M, 120°C, 20 mL/g)(a) (c) (b) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017    Figure 6. Analysis for Au recovered from sample 1 (a) SEM analysis for the Au coating, (b) EDX spectrum (c) Fine coatings of Au             Figure 7. Analysis for Au recovery from sample 2 (a) SEM analysis for the Au coating, (b) EDX spectrum (c) Fine coatings of Au      (a) (b) (c) (a) (b) (c) International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017    Figure 8. Analysis for Au recovery from sample 3 (a) SEM analysis for the Au coating, (b) EDX spectrum (c) Fine coatings of Au  CONCLUSIONS  Gold was recovered from computer memory boards, electronic processors and electronic pins and contacts performing selective base metals oxidation in (NH4)2S2O8 solutions. The process parameters affecting the base metals oxidation and the recovery of gold were studied following a systematic experimental design that included full factorial and central composite designs with response surface methodology. The optimum conditions for recovery of gold greater than 96% were established by the optimization of the process parameters. The optimization established reaction times of 5, 10 and 15 minutes for base metals oxidation. In those times, the oxidative partial dissolution of the substrate took place breaking the Au-Ni-Fe-Cu bonds, and fine coatings of non-leaching gold were recovered in a 99.36%, 99.43% and 96.23% from samples 1, 2 and 3 respectively. The development of a first order equation for the prediction of Au recovery from sample 1 explained the variability associated with the model. Nevertheless, second order models were tested using response surface methodology to analyze the oxidative conditions in samples 2 and 3. By using RSM, it was possible to minimize the (NH4)2S2O8 concentration (0.66 M), oxygen flow (0.6 L/min) and L/S ratio (20 mL/g) for the Au recovery from sample 2 and minimize the (NH4)2S2O8 concentration (0.88 M), temperature (120°C) and L/S ratio (20 mL/g) for the Au recovery from sample 3. The findings presented in this paper suggest that (NH4)2S2O8 can be used as an environmentally friendly methodology to recover gold from distinct types of e-waste by using optimum levels of (NH4)2S2O8 concentration, Oxygen flowrate, pressure, temperature and liquid/solid ratio. This methodology provides insights on how to reduce agent consumption by combined mechanical, thermal and chemical parameters that have a direct influence on the reaction rates.  ACKNOWLEDGEMENTS  The authors would like to sincerely acknowledge the national fund for the financing of science, technology and innovation (FRANCISCO JOSE DE CALDAS) and the University of Antioquia for the financial support to successfully develop this research. The authors would also like to thank the Colombian companies INSUMON S.A.S and LITO S.A.S for providing samples, staff and facilities for the experimental tests and the University of British Columbia - UBC for providing facilities for the characterization and kinetics test. REFERENCES  Alzate, A., López, M.E., Serna, C., 2016a. Recovery of gold from waste electrical and electronic equipment (WEEE) using ammonium persulfate. Waste Manag. 57, 113–120. doi:10.1016/j.wasman.2016.01.043 Alzate, A., López, M.E., Serna, C., 2015. Experimental Study on the Incidence of Various Factors in the Base Metals Leaching to Recover Gold from Electrical and Electronic Waste, in: Graber, T., International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  Taboada, M., Valenzuela, F. (Eds.), 7th International Seminar on Process Hydrometallurgy. Gecamin, Santiago, Chile. Alzate, A., López, M.E., Serna, C., Gonzalez, O., 2016b. Gold Recovery from Electronic Waste by Pressure Oxidation, in: Proceedings of the 2nd World Congress on Mechanical, Chemical, and Material Engineering (MCM’16), Budapest, Hungary – August 22 – 23, 2016. Budapest, pp. 1–7. doi:10.11159/mmme16.109 Bas. D, B. I., 2007. Modeling and optimization I: Usability of response surface methodology. Journal of food engineering 78, 836-845. Barbieri, L., Giovanardi, R., Lancellotti, I., Michelazzi, M., 2010. A new environmentally friendly process for the recovery of gold from electronic waste. Environ. Chem. Lett. 8, 171–178. doi:10.1007/s10311-009-0205-2 Birloaga, I., De Michelis, I., Ferella, F., Buzatu, M., Vegliò, F., 2013. Study on the influence of various factors in the hydrometallurgical processing of waste printed circuit boards for copper and gold recovery. Waste Manag. 33, 935–941. doi:10.1016/j.wasman.2013.01.003 Design-Expert., 2015. Design-Expert, Version 9.0.4.1. Stat-Ease Inc., 2021 East Hennepin Ave, Suite 480, Minneapolis, MN 55413. Ghosh, B., Ghosh, M.K., Parhi, P., Mukherjee, P.S., Mishra, B.K., 2015. Waste Printed Circuit Boards recycling: An extensive assessment of current status. J. Clean. Prod. 94, 5–19. doi:10.1016/j.jclepro.2015.02.024 Ha, V.H., Lee, J., Huynh, T.H., Jeong, J., Pandey, B.D., 2014. Optimizing the thiosulphate leaching of gold from printed circuit boards of discarded mobile phone. Hydrometallurgy 149, 118–126. doi:10.1016/j.hydromet.2014.07.007 Hadi, P., Xu, M., Lin, C.S.K., Hui, C.-W., McKay, G., 2015. Waste printed circuit board recycling techniques and product utilization. J. Hazard. Mater. 283, 234–243. doi:10.1016/j.jhazmat.2014.09.032 Hernandez, O., 2005. Persulfates. UNEP Publications, SIDS Initial Assessment Report For SIAM 20, Washington, pp. 1–192. Hyk, W., Kitka, K., 2017. Highly efficient and selective leaching of silver from electronic scrap in the base-activated persulfate – ammonia system. Waste Manag. doi:10.1016/j.wasman.2016.12.038 Jordão, H., Sousa, A.J., Carvalho, M.T., 2016. Optimization of wet shaking table process using response surface methodology applied to the separation of copper and aluminum from the fine fraction of shredder ELVs. Waste Manag. 48, 366–373. doi:10.1016/j.wasman.2015.10.006 Lee, C.-H., Tang, L.-W., Popuri, S.R., 2011. A study on the recycling of scrap integrated circuits by leaching. Waste Manag. Res. 29, 677–685. doi:10.1177/0734242X10380995 Montgomery, D., 2013. Design and analysis of experiments second edition. Mexico: Limusa Wiley. Montgomery. D., 2011. Engineering Statistics. Arizona State University: John Wiley & Sons, Inc Naseri Joda, N., Rashchi, F., 2012. Recovery of ultra fine grained silver and copper from PC board scraps. Sep. Purif. Technol. 92, 36–42. doi:10.1016/j.seppur.2012.03.022 Petter, P.M.H., Veit, H.M., Bernardes, A.M., 2014. Evaluation of gold and silver leaching from printed circuit board of cellphones. Waste Manag. 34, 475–482. doi:10.1016/j.wasman.2013.10.032 Sanyal, S., Ke, Q., Zhang, Y., Ngo, T., Carrell, J., Zhang, H., Dai, L.L., 2013. Understanding and optimizing delamination/recycling of printed circuit boards using a supercritical carbon dioxide process. J. Clean. Prod. 41, 174–178. doi:10.1016/j.jclepro.2012.10.011 Sarvar, M., Salarirad, M.M., Shabani, M.A., 2015. Characterization and mechanical separation of metals from computer Printed Circuit Boards (PCBs) based on mineral processing methods. Waste Manag. 45, 246–257. doi:10.1016/j.wasman.2015.06.020 Sharma, J., Mishra, I.M., Kumar, V., 2015. Degradation and mineralization of Bisphenol A (BPA) in aqueous solution using advanced oxidation processes:UV/H2O2 and UV/S2O82- oxidation systems. J. Environ. Manage. 156, 266–275. doi:10.1016/j.jenvman.2015.03.048 Syed, S., 2012. Recovery of gold from secondary sources—A review. Hydrometallurgy 115–116, 30–51. doi:10.1016/j.hydromet.2011.12.012 Syed, S., 2006. A green technology for recovery of gold from non-metallic secondary sources. Hydrometallurgy 82, 48–53. doi:10.1016/j.hydromet.2006.01.004 Tuncuk, A., Stazi, V., Akcil, A., Yazici, E.Y., Deveci, H., 2012. Aqueous metal recovery techniques from International Journal of Mineral Processing/Minerals Engineering-special issue after COM 2017  e-scrap: Hydrometallurgy in recycling. Miner. Eng. 25, 28–37. doi:10.1016/j.mineng.2011.09.019 Turan, M.D., Arslano, H., Altundo, H.S., 2015. Optimization of the leaching conditions of chalcopyrite concentrate using ammonium persulfate in an autoclave system. J. Taiwan Inst. Chem. Eng. 50, 49–55. doi:10.1016/j.jtice.2014.12.009 Walpole. R, M. R., 2012. Probability & statistics for engineers & scientists. Boston: Prentice Hall. Xiu, F.-R., Qi, Y., Zhang, F.-S., 2015. Leaching of Au, Ag, and Pd from waste printed circuit boards of mobile phone by iodide lixiviant after supercritical water pre-treatment. Waste Manag. 41, 134–141. doi:10.1016/j.wasman.2015.02.020 Yazici, E.Y., Deveci, H., 2014. Ferric sulphate leaching of metals from waste printed circuit boards. Int. J. Miner. Process. 133, 39–45. doi:10.1016/j.minpro.2014.09.015 Yazici, E.Y., Deveci, H., 2013. Extraction of metals from waste printed circuit boards (WPCBs) in H 2SO4-CuSO4-NaCl solutions. Hydrometallurgy 139, 30–38. doi:10.1016/j.hydromet.2013.06.018 Zhang, Y., Liu, S., Xie, H., Zeng, X., Li, J., 2012. Current Status on Leaching Precious Metals from Waste Printed Circuit Boards. Procedia Environ. Sci. 16, 560–568. doi:10.1016/j.proenv.2012.10.077  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0347214/manifest

Comment

Related Items